公开(公告)号：US20170139568A1

公开(公告)日：2017-05-18

申请号：US14942837

申请日：2015-11-16

申请人： Steiman Itani ,  Yu-Hsiang Chen ,  Mohamed Nabil Hajj Chehade ,  Allen Yang Yang

发明人： Steiman Itani ,  Yu-Hsiang Chen ,  Mohamed Nabil Hajj Chehade ,  Allen Yang Yang

IPC分类号： G06F3/0481 ,  G06F3/01 ,  G02B27/01 ,  G06F3/041 ,  G06K9/00

CPC分类号： G06F3/04812 ,  G02B27/017 ,  G02B2027/014 ,  G06F3/04883 ,  G06F2203/04808 ,  G06K9/00355 ,  G06K9/6253

摘要： To prompt input and provide feedback on input to a user with an interface, inputs and graphical cursors associated with those inputs are defined. Each input may have several forms such as base, hover, engaged, completed, and error. User input is anticipated. The base form of the anticipated input cursor is displayed to prompt the user for the anticipated input. If user hover is detected that matches anticipated input, the hover form is displayed to confirm the match to the user. If user input is detected that matches anticipated input, the engaged form is displayed as confirmation. If user input is completed that matches anticipated input, the completed form is displayed as confirmation. If user hover or input does not match anticipated input, the error form is displayed to indicate mismatch. Not all cursors must have all forms, and some cursors may have multiples of some forms.

公开(公告)号：US08406525B2

公开(公告)日：2013-03-26

申请号：US12865639

申请日：2009-01-29

申请人： Yi Ma ,  Allen Yang Yang ,  John Norbert Wright ,  Andrew William Wagner

发明人： Yi Ma ,  Allen Yang Yang ,  John Norbert Wright ,  Andrew William Wagner

IPC分类号： G06K9/66

CPC分类号： G06K9/00275 ,  G06K9/6232 ,  G06K9/627

摘要： A method is disclosed for recognition of high-dimensional data in the presence of occlusion, including: receiving a target data that includes an occlusion and is of an unknown class, wherein the target data includes a known object; sampling a plurality of training data files comprising a plurality of distinct classes of the same object as that of the target data; and identifying the class of the target data through linear superposition of the sampled training data files using l1 minimization, wherein a linear superposition with a sparsest number of coefficients is used to identify the class of the target data.

摘要翻译： 公开了一种用于在存在遮挡的情况下识别高维数据的方法，包括：接收包括闭塞并且是未知类的目标数据，其中所述目标数据包括已知对象; 对包含与所述目标数据相同的对象的多个不同类别的多个训练数据文件进行采样; 以及使用l1最小化通过采样的训练数据文件的线性叠加来识别目标数据的类别，其中使用具有最少数量的系数的线性叠加来标识目标数据的类别。

公开(公告)号：US20110064302A1

公开(公告)日：2011-03-17

申请号：US12865639

申请日：2009-01-29

申请人： Yi Ma ,  Allen Yang Yang ,  John Norbert Wright ,  Andrew William Wagner

发明人： Yi Ma ,  Allen Yang Yang ,  John Norbert Wright ,  Andrew William Wagner

IPC分类号： G06K9/62 ,  G10L15/08 ,  G06N5/02

CPC分类号： G06K9/00275 ,  G06K9/6232 ,  G06K9/627

摘要： A method is disclosed for recognition of high-dimensional data in the presence of occlusion, including: receiving a target data that includes an occlusion and is of an unknown class, wherein the target data includes a known object; sampling a plurality of training data files comprising a plurality of distinct classes of the same object as that of the target data; and identifying the class of the target data through linear superposition of the sampled training data files using l1 minimization, wherein a linear superposition with a sparsest number of coefficients is used to identify the class of the target data.

摘要翻译： 公开了一种用于在存在遮挡的情况下识别高维数据的方法，包括：接收包括闭塞并且是未知类的目标数据，其中所述目标数据包括已知对象; 对包含与所述目标数据相同的对象的多个不同类别的多个训练数据文件进行采样; 以及使用l1最小化通过采样的训练数据文件的线性叠加来识别目标数据的类别，其中使用具有最少数量的系数的线性叠加来标识目标数据的类别。